cached-path
smart_open
pypdf>=5.2.0
pypdfium2
cryptography
lingua-language-detector
Pillow
ftfy
bleach
markdown2
filelock
orjson
requests
zstandard
boto3
httpx
torch>=2.7.0
transformers==4.52.4
img2pdf
beaker-py

[bench]
tinyhost
fuzzysearch
rapidfuzz
sequence_align
syntok
openai
google-genai
playwright
mistralai
lxml
flask

[dev]
ruff
mypy
black
isort
pytest
pytest-sphinx
pytest-cov
pytest-asyncio
twine>=1.11.0
build
setuptools
wheel
Sphinx<7.1.0,>=4.3.0
furo==2023.7.26
myst-parser<2.1,>=1.0
sphinx-copybutton==0.5.2
sphinx-autobuild==2021.3.14
sphinx-autodoc-typehints==1.23.3
packaging
necessary
peft
datasets
omegaconf
spacy

[elo]
numpy
scipy
pandas
matplotlib

[gpu]
vllm==0.9.2

[train]
torch
torchvision
accelerate
datasets
peft
wandb
omegaconf
s3fs
necessary
einops
